This article was downloaded by: [Ataturk University]On: 22 October 2014, At: 06:14Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Desalination and Water TreatmentPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/tdwt20

Prolonged reuse of domestic wastewater aftermembrane bioreactor treatmentOkan Tarık Komeslia, Melis Muzb, Selcen Akb & Celal Ferdi Gökçayb

a Department of Environmental Engineering, Ataturk University, Erzurum 25250, Turkeyb Department of Environmental Engineering, Middle East Technical University, Ankara 06800,TurkeyPublished online: 07 Jul 2014.

To cite this article: Okan Tarık Komesli, Melis Muz, Selcen Ak & Celal Ferdi Gökçay (2014): Prolonged reuse of domesticwastewater after membrane bioreactor treatment, Desalination and Water Treatment, DOI: 10.1080/19443994.2014.934107

To link to this article: http://dx.doi.org/10.1080/19443994.2014.934107

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Prolonged reuse of domestic wastewater after membrane bioreactor treatment

Okan Tarık Komeslia,*, Melis Muzb, Selcen Akb, Celal Ferdi Gokcayb

aDepartment of Environmental Engineering, Ataturk University, Erzurum 25250, Turkey, email: okan.komesli@gmail.combDepartment of Environmental Engineering, Middle East Technical University, Ankara 06800, Turkey,emails: melis.muz@ufz.de (M. Muz), mselcen@metu.edu.tr (S. Ak), cfgokcay@metu.edu.tr (C.F. Gokcay)

Received 11 October 2013; Accepted 13 January 2014

ABSTRACT

In this study, experience in a nine-year operation of a full-scale 2000 PE vacuum rotatingmembrane bioreactor having a submerged flat-type membrane module having pore size of0.038 μm and a total surface area of 540m2 is discussed. The plant was designed to treatand reuse raw wastewater collected from dormitories and the academic village at METUcampus. Throughout the study, 99.99% BOD5 and above 95% COD removals were achievedmost of the time. Moreover, turbidity was consistently measured below 1NTU and around6–7 log coliform removals were achieved with less than 1 coliform/100mL in the effluentsmost of the time, except for the leakage from the bearings. During the study, energy con-sumption by the plant was also analyzed by routinely measuring energy consumption indifferent parts of the plant. Consumption was analyzed in two parts. Energy consumed bythe blower supplying aeration to the biological treatment tank was monitored separatelyfrom the rest of the plant. Except for the periods when problems have occurred duringoperation, the total energy consumption of the system was variable between 1.1 and 2.53kWh/m3, averaging around 2 kWh/m3. The main problems encountered during operationwere poor floc formation and dispersed growth, and sludge deposition between the mem-brane plates and mechanical malfunction of the bearing seals. The treated wastewaters werestored and used for the irrigation of METU Technopolis lawns.

Keywords: Wastewater treatment; Membrane bioreactor; Prolonged; Reuse

1. Introduction

The world population is estimated to grow dra-matically from 2004 to 2020 with accompanying scar-city of clean waters [1]. Due to this increase, cleanwater resources are becoming increasingly scarce inmany areas of the world [2,3]. In order to obtain newfresh water resources, newer technologies have been

investigated. One of the innovative activated sludgeprocesses for wastewater reuse is membrane bioreac-tor (MBR), which employs membrane filtration insteadof secondary clarifiers to achieve biomass separation[4–7]. Although filters have been use since early 1960sas filtration devices, their usage as a means of waterand wastewater treatment goes back only few decades[8]. MBR technology is considered superior over theconventional biological systems in that it produces farbetter-quality effluents. In MBRs, hydraulic retention

*Corresponding author.

Presented at the International Conference WIN4Life, 19–21 September 2013, Tinos Island, Greece

1944-3994/1944-3986 � 2014 Balaban Desalination Publications. All rights reserved.

Desalination and Water Treatment (2014) 1–8

www.deswater.com

doi: 10.1080/19443994.2014.934107

Dow

nloa

ded

by [

Ata

turk

Uni

vers

ity]

at 0

6:14

22

Oct

ober

201

4

time (HRT) is independent of the sludge retentiontime, SRT [9], and truly infinite SRT is achievable inthese systems [10]. In addition to producing sparklingclear effluents, MBRs are efficient in concentratingmixed liquor suspended solids (MLSS) to remarkablyhigh levels, thereby reducing plant footprint [11].Apart from the many advantages of MBRs, the highinvestment cost was the main drawback of this tech-nology by the end of 1990s. Due to the advancementof the polymer industry, membrane costs declinedrapidly after the turn of the twentieth century [6,12]and MBR plants became widespread. Moreover, inearly 1990s, MBR plants were mostly being con-structed in external configuration, where membranemodules were located outside the bioreactor and bio-mass was recirculated between the filtration unit andthe reactor. Although the external membrane alloweda better access for the cleaning and reducing the mem-brane fouling, the operation cost was very high andnot applicable for the treatment and reuse of munici-pal wastewater. External type of membrane is mostlyused for industrial wastewater treatment. Owing totheir high electricity consumption, submerged MBRsbecame the method of choice in municipal wastewatertreatment after the mid-1990s [13]. This developmenthas led to wider application of MBR plants in theworld, as of the year 2000. Therefore, after 1990s, sub-merged MBR systems were developed for the treat-ment and reuse of municipal wastewaters. In thisconfiguration, membrane modules were directlyimmersed in the activated sludge tank to decrease theelectricity utilization of the system. In addition, thereare many advantages of submerged systems, includingsimple design and higher hydraulic efficiencies, thanexternal design. The transmembrane pressure (TMP)in submerged system is very low, around 0.3 bar, com-pared with external configuration, from 1 to 4 bar [14].

Wastewater reuse is done mainly for irrigation, toi-let flushing, and some cleaning actions. In order toreuse wastewaters after treatment in irrigation, some

regulations were set up to minimize the health risks ofhuman exposure to the pathogens. In Turkish WaterPollution Control Regulation, “Technical Aspects Bul-letin,” was published on 4 September 1988 and consid-ered during the study [15].

In this study, a nine-year operation performance ofa full-scale vacuum rotating membrane (VRM) reactor,operating in METU campus, Ankara, Turkey, is ana-lyzed.

2. Materials and methods

2.1. The VRM MBR system at METU

A full-scale MBR, referred to as VRM, located inMETU Campus, Ankara, was used in this study. Asseen in Fig. 1, the plant consists of two tanks and theperipheral equipment. A partitioning wall between thetanks separates the two. However, the two tanks areconnected by five orifices, each controlled manually,located at the bottom of the partitioning wall. Theworking volume of the first tank is 85m3, and is usedfor the aeration of the biological sludge. The secondtank is about 23m3 in volume and is used to housethe VRM unit. Wastewater from dormitories and aca-demic village is first collected in a 10m3 holding tankand then pumped to the treatment plant located 50maway from the storage tank at 15m elevation. A 4 cmcoarse screen is located at the inlet to the storage tank.Wastewater is screened through a screw type finescreen having 3mm openings, Rotomat Ro9, producedby Huber A.G. at the entry to the aeration tank.

Membrane diffusers for aeration are placed at thebottom of the aeration tank. On the VRM unit, whosepicture is given in Fig. 2, flat plates are seated on adrum-like filter holder produced by Huber A.G. Thefilter holder is in continuous rotation driven by amotor. The rotation speed is 2.5 rpm. Suction isapplied to the plate modules from inside via six radialhoses connecting the suction pump to the suction

Fig. 1. VRM plant in METU.

2 O.T. Komesli et al. / Desalination and Water Treatment

Dow

nloa

ded

by [

Ata

turk

Uni

vers

ity]

at 0

6:14

22

Oct

ober

201

4

tubing on the membrane modules. A cross flow isapplied over the plates by coarse aeration throughtwo diffusers located under the plates at the center ofthe rotating drum. Membrane plates are separated15–20mm apart. Solids retention time (SRT) employedduring this study were between 10 and 150 d, andHRT, was between 15 and 24 h. The typical flux ratesemployed were 8.3–13 L/hm2 and the maximumattainable was 15 L/hm2. A picture of the VRM unitat the factory shop floor and a membrane unit is givenin Fig. 2. A portion of the mixed liquor was continu-ously being recirculated from the VRM tank to theaeration tank. Its ratio to the inflow was 3. Systemwas initially operated in 10min cycles with 8min vac-uum and 2min relaxation. This was later changed to4min vacuum and 1min relaxation, which alsoworked equally well. During relexation period,although vacuum pump was stopped, aeration ofmembran module and rotation were continued.

2.2. In situ, on-line measurements

These measurements included TMP, temperature(T), dissolved oxygen (DO), and pumping flows. Tem-perature and DO were measured online by a JumodTrans O2-01 DO and temperature meter. The DOconcentration in the aeration tank was controlled bymeans of a DO probe and a process control consolelinked to the blower, whereas temperature was onlyrecorded. The probe was submerged in the aerationtank, and data obtained were transmitted to the PLCmodule located inside the process control console.Pumped flow was measured by a Siemens ultrasonicflow meter and TMP was measured by a transducerplaced at the inlet of the vacuum pump. The onlinemeasurements were collected continuously during

operation and stored in a Micromec data logger anddownloaded at constant intervals.

2.3. Laboratory analysis

The COD, BOD5, total suspended solid (TSS), tur-bidity, and conductivity were routinely analyzed inthe laboratory in duplicates. COD was measuredusing Hach Dr 2000 Model Spectrophotometer. HachCOD reagent (Cat No. 21259-51) for COD was used.The BOD5 of the influent and effluent was measuredaccording to Standard Method (5210B) [16]. TSS wasanalyzed according to Standard Methods (2540B) [16].Standard Methods 9222B and 9222D were used for theanalyses of total and fecal coliform. A Hach 2100 Nmodel turbidimeter was used to determine the influ-ent and effluent turbidities. A YSI 33 model conductiv-ity and salinity meter was used for the measurementof conductivity. The electricity consumption was mea-sured directly from the control panel of VRM plant.

3. Result and discussion

Full-scale VRM plant was successfully operated formore than nine years, and is still in operation in theMETU Campus, Ankara. The operation was com-menced in May 2004 with the transfer of activatedsludge from Ankara Wastewater Treatment Plant, hav-ing 3 g/L MLSS. Initially, the plant was operated at7.5 m3/h and 13.8 L/m2 h flux rates. The HRT of thesystem was arranged to 14–22 h and the plant opera-tion regime was initially set as 8min suction followedby 2min relaxation without suction. This was laterchanged to 4min suction and 1min relaxation, whichworked equally well. At the end of 5th year of opera-tion, about 50m2 of membrane area was damaged and

Fig. 2. The VRM membrane holder and a membrane module cassette.

O.T. Komesli et al. / Desalination and Water Treatment 3

Dow

nloa

ded

by [

Ata

turk

Uni

vers

ity]

at 0

6:14

22

Oct

ober

201

4

flux was increased to about 16 L/m2 h to achieve thesame flowrate.

During the entire nine years of the study period,the SRT was arranged from 10 to 150 d. At the begin-ning of the study, MLSS concentration was set 150 dwithout disposal of sludge. During this period, MLSSconcentration in the membrane chamber has increasedfrom 8 to 21 g/L. After this period, sludge was dis-posed not only to decrease the sludge concentration inthe aeration tank but also to increase the active bio-mass population in the reactor. The highest sludgeconcentration reached during this period was 23 g/L,whereupon sludge was disposed to decrease MLSSconcentration. The MLVSS/MLSS was between 0.50and 0.78 during the study.

During the study, measurements of COD in influ-ent and effluent were carried out routinely to under-stand assess efficiency of the system. The plant was sorobust that daily fluctuations in influent COD had noeffect over the treatment efficiency, even at SRTs aslow as 10 d, as shown in Fig. 3.

As seen in Fig. 3, the peak COD was observed atnoon and the lowest in the early morning. The aver-age influent COD concentration was 426mg/L andeffluent COD concentration was steady at 19mg/L.This indicated an average of 96% COD removal dur-ing the day. Throughout the study, the influent CODwas measured between 240 and 1,200mg/L. DO con-centration in the aeration tank was variable between0.1 and 4mg/L. The plant still produced clear effluentwhen DO was so low. The highest influent COD wasmeasured in summer of 2008 during a drought.Effluent COD concentration was still consistentlybelow 60mg/L, when the highest influent concentra-tion was met and DO concentration was around 0.1mg/L. This lowest oxygen concentration could beexplained by the oxygen consumption during theorganics removal. In addition, higher concentration ofMLSS during that period was over 12 g/L in aerationtank. In other times, during normal operation, the

effluent COD was mostly around 20mg/L. A checkon the effluent COD indicated that it is not biodegrad-able, probably originating from the inert influent CODand/or the soluble inert microbial products. Therecorded influent and effluent COD concentrations inthe course of operation between January 2004 andAugust 2012 are presented in Fig. 4, along with thecorresponding COD removals.

The most important parameters for the assessmentof reuse of wastewaters for irrigation are turbidity andcoliform counts. Fecal and total coliform organismsare taken as good indicators of water quality. Particu-larly, fecal coliforms are true indicators of human ori-gin, whereas total coliforms may sometimes bemisleading as these may also be free living in nature.Turkish reuse standard, just like the US and Israelstandards, calls for less than 2 NTU and 2/100 fecalcoliform count for unrestricted reuse. Typical fecal colibacilli counts attained at the METU VRM treatmentplant are depicted in Fig. 5. From this figure, it is seenthat less than 2 coli bacilli/100 ml could be achievedin the effluents. Since the pore size of the membraneis about 0.038 μm and the size of bacteria is about1 μm, which is about 25 times bigger than pore size, itwas not possible to pass bacteria from the membranefilter. Most of the times, over 90% of the samples, thenumber of fecal coliform in the effluent was zero.However, in some situations, due to contaminationwhen taking samples, it could be possible to measure

Fig. 3. Influent COD fluctuation during the day.Fig. 4. Typical influent and effluent COD values with cor-responding COD removal percentages.

4 O.T. Komesli et al. / Desalination and Water Treatment

Dow

nloa

ded

by [

Ata

turk

Uni

vers

ity]

at 0

6:14

22

Oct

ober

201

4

FC in the effluent. It was also clearly seen from theturbidity measurements that it was not any damage inmembrane modules. During this nine-year period ofthe operation, the number of fecal coliform in theeffluent was less than 1 in 90% of effluent samples.Approximately, 7 log reductions in fecal coliformcounts and 5–6 log reductions in total coliform countshave been constantly achieved.

Turbidity of the influent was between 115 and 210NTU (Nephelometric Turbidity Units), while typicallythis ranged between 0.1 and 1.1 NTU, as shown inFig. 6. Turbidity of the water was around 0.5 duringthis period.

Influent conductivity was measured typicallybetween 1,350 and 1,450 μmho/cm. There was very littledecrease in the effluent conductivity through treatment,which ranged between 1,250 and 1,350 μmho/cm. Con-ductivity of the tap water was around 1,000 μmho/cm.A picture of the sprinkle irrigation employed at METUTechnopolis lawns is shown in Fig. 7.

3.1. Rheology of VRM sludge

Rheology of sludge is extremely important duringthe operation of the VRM plant since highly viscoussludge tends to block spaces between membraneplates. This became even more important when sludgeconcentration in the VRM tank has increased to over20 g/L when the plant had to be operated in extendedaeration mode without excess sludge wastage. Therheology of sludge increases dramatically as the MLSSconcentration in the aeration tank increases. Thisbecomes detrimental when viscosity goes beyond 20cP. In fact, 10% of the filter media was lost when try-ing to remove a blockage of inter-membrane spaces.During blockage, effluent flux tends to decline andTMP increases, and steady TMP below −350mbar isindicative of chemical cleanup. In conventional acti-vated sludge systems, sludge mostly behaves non-Newtonian and pseudo plastic [17]. Similarly, MBRsludge is reported to behave non-Newtonian andpseudo plastic [18] in character, whereas in VRMplant, rheological characteristics of sludge in the filterchamber are plastic. A truly plastic sludge should bebetter and more strongly packed into interplate spacescausing strong blockages. The viscosity observationson this plant are already published by Komesli andGokcay [19].

3.2. Energy consumption and cost calculations

Energy consumption is an important aspect inwastewater treatment plant, which, sometimes, couldaffect the feasibility of a treatment method. In order tooptimize the energy utilization of VRM plant, the

Fig. 5. Typical numbers of fecal coliforms recorded in theeffluent.

Fig. 6. Effluent water turbidities.

Fig. 7. Irrigation of lawns with reuse water at METUTechnopolis.

O.T. Komesli et al. / Desalination and Water Treatment 5

Dow

nloa

ded

by [

Ata

turk

Uni

vers

ity]

at 0

6:14

22

Oct

ober

201

4

electricity consumption during the study was investi-gated. In VRM plant, energy consumed was dividedinto two parts: fixed and variable energy consump-tion. Fixed electricity utilization includes power con-sumption by rotation, blower for coarse aeration,control panel, and fine screen. Variable energy utiliza-tion was due to inflow, permeate and recirculationpumps, and consumption by the blower for activatedsludge. The energy utilization by each device percubic meter of treated effluent, when treating 6m3/h,is given in Fig. 8.

It is clearly seen from Fig. 8 that almost half of theenergy was consumed by fine aeration, whereas thehighest energy consumption is attributed to the coarseaeration in the literature by about 2 kWh/m3 [20]. Thismay be attributed to the fact that, in static plants, thecross flow is maintained by installing a series of dif-fusers to cover the entire bottom of the membraneplates, whereas in VRM, only two diffuser pipes areplaced at the center of the membrane holder unit andcoarse aeration is maintained from this point, although

membranes revolve around the diffusers to exposethem alternatively to the cross flow. Since energyutilization of a treatment plant depends on the capac-ity of the plant, it is not possible to completely com-pare the energy utilization of different capacities ofthe plants. However, due to Fenu et al., course bubbleaeration was the largest energy consumption device instatic MBR plant [14]. In another study conducted byKrzeminski et al., three different configurations offull-scale MBR plants were compared. However, thecapacity of each configuration was different and10–100 times bigger than our plant. In those plants,the energy utilization was from 0.6 to 1.8 kWh/m3

[21]. In this study, the largest energy utilization devicewas also coarse bubble aeration from 36 to 64 of allthe energy consumption. In another study by Uedaet al., MBR plant with 4m2 membrane surface areawas used and the energy consumption was 70, whichwas very high and was not meant to be compared bythis study [22].

The total energy consumption by the plant chan-ged appreciably in time since the commissioning ofthe membrane plant, i.e. from 1.1 to 2.53 kWh/m3

(permeate), due to wear on the blowers and the per-meate pump. Our observation showed that underclose control and optimum maintenance, the cost ofoperation may be reduced to 1.4 kWh/m3. Taking thedepreciation period of the membrane as 10 years, atthe most, and assuming 140m3/d uninterrupted flowfor 10 years, and the most recent quote for 540m2

membrane as € 80,000, the depreciation cost of mem-branes corresponding to 1m3 of treated wastewaterworks out as €0.156. Therefore, the realistic cost of 1m3 reuse of wastewater is 2.14 kWh/m3 and €0.156 forthe membrane depreciation. This figure may bereduced to 1.4 kWh/m3 and €0.156/m3, at the idealcase. Considering the selling price of electricity, it is

Fig. 8. Normalized energy consumption by each devicemaking up the VRM plant.

Fig. 9. Sludge layer on membrane plates and filled inter-plate spaces.

6 O.T. Komesli et al. / Desalination and Water Treatment

Dow

nloa

ded

by [

Ata

turk

Uni

vers

ity]

at 0

6:14

22

Oct

ober

201

4

€cents 9.5 in Turkey in 2012, and the complete cost of1m3 reuse water is calculated as €cents 35/m3. This isaround 1/10th of the selling price of water by themunicipality in Ankara, which is currently €3/m3.

3.3. Operational problems

During the initial years, several mechanical prob-lems have occurred in the plant. Most serious wasthe leakage of the central bearing seal. The problemwas realized when mixed liquor in the filter chamberpenetrated into the vacuum area causing increase ineffluent NTU. The problem was rectified by HuberA.G. technicians by adopting a new bearing design.Since then, VRM is operating flawless with its regu-larly maintained bearings. The other important prob-lem to be aware of is sludge accumulation betweenthe membrane plates as shown in Fig. 9. Occasionalchemical cleaning of the membrane modules with0.5% NaOCl− for 5–10 h, as per required, frees clog-ging in mild cases. Sludge accumulation was noticedwhen MLSS concentration approached 20 g/L, but sit-uation could be rectified by lowering MLSS and byprolonged operation of the VRM without suction andwashing plate interspaces by pressurized water.Dismantling or handling aged membrane plates isextremely risky as they get brittle in time and touch-ing by hand causes punctures. Maintaining low MLSSconcentrations, i.e. below 12 g/L or 20 cP, and 2–3times chemical cleansing per year when TMPpersisted to stay below −300mbar prevented cloggingof the METU plant.

4. Conclusion

During nine years of operation of a full-scale VRMplant, it has been demonstrated that treated effluentscould successfully be used for the irrigation of sensi-tive lawns at a reasonable cost. The COD removal wasconsistently greater than 90–95%, and the COD in theeffluent was below 20mg/L for most of the samples.Turbidity and bacteria removals were excellent in thissystem, producing effluents with less than 1NTU andclose to zero fecal coliform count all the time. Block-age of inter-plate spaces, which is a form of fouling,might cause irreversible damage to the membranes.Hence, keeping an MLSS concentration below 12 g/Land a viscosity reading below 20 cP for the mixedliquor are recommended to avoid this to happen.Energy consumption of the system is lower than otherMBR systems, which are stable. Rotation movementis not only for reducing the energy but also for

increasing the cross flow in the tank. During nineyears operation, membrane modules have not beenchanged and are continuing to filtration.

Acknowledgments

This research was funded by TUBITAK project no.1008Y272. We thank Huber A.G. for the donation ofthe VRM unit and METU Technopolis for theircontributions.

References

[1] Anonymous, EPA, Guidelines for Water Reuse, U.S.Environmental Protection Agency, EPA/625/R-04/108, Washington, DC, 2004.

[2] G.K. Pearce, UF/MF pre-treatment to RO in seawaterand wastewater reuse applications: A comparison ofenergy costs, Desalination 222 (2008) 66–73.

[3] G. Tchobanoglous, F.L. Burton, Wastewater Engineer-ing: Treatment and Reuse, fourth ed., Medcalf andEddy, Singapore, 2003.

[4] B. Gunder, The membrane coupled-activated sludgeprocess in municipal wastewater treatment. Technomicpublishing company, Lancaster, 2001.

[5] J.A. Howell, Future of membranes and membranereactors in green technologies and for water reuse,Desalination 162 (2004) 1–11.

[6] T. Jiang, M.D. Kennedy, C.K. Yoo, I. Nopens, W. vander Meer, H. Futselaar, J.C. Schippers, P.A. Vanrolleg-hem, Controlling submicron particle deposition in aside-stream membrane bioreactor: A theoretical hydro-dynamic modelling approach incorporating energyconsumption, J. Membr. Sci. 297 (2007) 141–151.

[7] A. Dhouib, N. Hdiji, I. Hassaıri, S. Sayadi, Large scaleapplication of membrane bioreactor technology for thetreatment and reuse of an anionic surfactant wastewa-ter, Process Biochem. 40 (2005) 2715–2720.

[8] N. Scharnagl, U. Bunse, K.V. Peinemann, Recycling ofwashing waters from bottle cleaning machines usingmembranes, Desalination 131 (2000) 55–63.

[9] S.H. Yoon, Important operational parameters of mem-brane bioreactor-sludge disintegration (MBR-SD) sys-tem for zero excess sludge production, Water Res. 37(2003) 1921–1931.

[10] C. Visvanathan, R.B. Aim, K. Parameshwaran, Mem-brane separation bioreactors for wastewater treatment,Crit. Rev. Environ. Sci. Technol. 30(1) (2000) 1–48.

[11] O.T. Komesli, K. Teschner, W. Hegemann, C.F.Gokcay, Vacuum membrane applications in domesticwastewater reuse, Desalination 215 (2007) 22–28.

[12] S. Judd, The MBR Book: Principles and Applicationsof Membrane Bioreactors, Elsevier, Boston, MA, 2006.

[13] S. Zhang, R. van Houten, D.H. Eikelboom, H.Doddema, Z. Jiang, Y. Fan, J. Wang, Sewage treatmentby a low energy membrane bioreactor, Bioresour.Technol. 90 (2003) 185–192.

[14] A. Fenu, J. Roels, T. Wambecq, K. De Gussem, C.Thoeye, G. De Gueldre, B. Van De Steene, Energyaudit of a full scale MBR system, Desalination 262(2010) 121–128.

O.T. Komesli et al. / Desalination and Water Treatment 7

Dow

nloa

ded

by [

Ata

turk

Uni

vers

ity]

at 0

6:14

22

Oct

ober

201

4

[15] Turkish Standard Institute, TS266, Water intended forhuman consumption, 1997.

[16] Standard Methods for the Examination of Water andWastewater, twentieth ed., APHA American PublicHealth Association, Washington, DC, 2001.

[17] P.A. Vesilind, Treatment and Disposal of WastewaterSludges, revised ed., Ann Arbor Science Publishers,Ann Arbor, MI, 1980.

[18] G. Laera, C. Giordano, A. Pollice, D. Saturno, G. Min-inni, Membrane bioreactor sludge rheology at differ-ent solid retention times, Water Res. 41(18) (2007)4197–4203.

[19] O.T. Komesli, C.F. Gokcay, Investigation of sludge vis-cosity and its effects on the performance of a vacuum

rotation membrane bioreactor, Environ. Technol. 35(5)(2014) 645–652.

[20] J.A. Gil, L. Tua, A. Rueda, B. Montano, M. Rodrıguez,D. Prats, Monitoring and analysis of the energy costof an MBR, Desalination 250 (2010) 997–1001.

[21] P. Krzeminski, W. Langhorst, P. Schyns, D. de Vente,R. Van den Broeck, I.Y. Smets, J.F.M. Van Impe, J.H.J.M.van der Graaf, J.B. van Lier, The optimal MBR configu-ration: Hybrid versus stand-alone—Comparisonbetween three full-scale MBRs treating municipalwastewater, Desalination 284 (2012) 341–348.

[22] T. Ueda, K. Hata, Y. Kikuoka, O. Seino, Effects of aer-ation on suction pressure in a submerged membranebioreactor, Water Res. 31(3) (1997) 489–494.

8 O.T. Komesli et al. / Desalination and Water Treatment

Dow

nloa

ded

by [

Ata

turk

Uni

vers

ity]

at 0

6:14

22

Oct

ober

201

4